Editor’s Note: I made this list off the top of my head and have surely missed awesome things people have done this year. If you know of some, you should make your own list or add it to the comments! I have also avoided talking about stuff I worked on or that people here at Hopkins are doing because this post is supposed to be about other people’s awesome stuff. I wrote this post because a blog often feels like a place to complain, but we started Simply Stats as a place to be pumped up about the stuff people were doing with data.
- I emailed Hadley Wickham about some trouble we were having memory profiling. He wrote back immediately, then wrote an R package, then wrote this awesome guide. That guy is ridiculous.
- Jared Horvath wrote this incredibly well-written and compelling argument for the scientific system that has given us a wide range of discoveries.
- Yuwen Liu and colleagues wrote this really interesting paper on power for RNA-seq studies comparing biological replicates and sequencing depth. Shows pretty conclusively to go for more replicates (music to a statisticians ears!).
- Yoav Benjamini and Yotam Hechtlingler wrote an amazing discussion of the paper we wrote about the science-wise false discovery rate. It contributes new ideas about estimation/control in that context.
- Sherri Rose wrote a fascinating article about statistician’s role in big data. One thing I really liked was this line: “This may require implementing commonly used methods, developing a new method, or integrating techniques from other fields to answer our problem.” I really like the idea that integrating and applying standard methods in new and creative ways can be viewed as a statistical contribution.
- Karl Broman gave his now legendary talk (part1/part2) on statistical graphics that I think should be required viewing for anyone who will ever plot data on a Google Hangout with the Iowa State data viz crowd. They had some technical difficulties during the broadcast so Karl B. took it down. Join me in begging him to put it back up again despited the warts.
- Everything Thomas Lumley wrote on notstatschat, I follow that blog super closely. I love this scrunchable poster he pointed to and this post on Statins and the Causal Markov property.
- I wish I could take Joe Blitzstein’s data science class. Particularly check out the reading list, which I think is excellent.
- Lev Muchik, Sinan Aral, and Sean Taylor brought the randomized control trial to social influence bias on a massive scale. I love how RCT are finding their ways into the new, sexy areas.
- Genevera Allen taught a congressman about statistical brain mapping and holy crap he talked about it on the floor of the house.
- Lior Pachter starting mixing it up on his blog. I don’t necessarily agree with all of his posts but it is hard to deny the influence that his posts have had on real science. I definitely read it regularly.
- Marie Davidian, President of the ASA, has been on a tear this year, doing tons of cool stuff, including landing the big fish, Nate Silver, for JSM. Super impressive to watch the energy. I’m also really excited to see what Bin Yu works on this year as president of IMS.
- The Stats 2013 crowd has done a ridiculously good job of getting the word out about statistics this year. I keep seeing statistics pop up in places like the WSJ, which warms my heart.
- One way I judge a paper is by how angry/jealous I am that I didn’t think of or write that paper. This paper on the reproducibility of RNA-seq experiments was so good I was seeing red. I’ll be reading everything that Tuuli Lappalainen’s new group at the New York Genome Center writes.
- Hector Corrada Bravo and the crowd at UMD wrote this paper about differential abundance in microbial communities that also made me crazy jealous. Just such a good idea done so well.
- Chad Myers and Curtis Huttenhower continue to absolutely tear it up on networks and microbiome stuff. Just stop guys, you are making the rest of us look bad…
- I don’t want to go to Stanford I want to go to Johns Hopkins.
- Ramnath keeps Ramnathing (def. to build incredible things at a speed that we can’t keep up with by repurposing old tools in the most creative way possible) with rCharts.
- Neo Chung and John Storey invented the jackstraw for testing the association between measured variables and principal components. It is an awesome idea and a descriptive name.
- I wasn’t at Bioc 2013, but I heard from two people who I highly respect and it takes a lot to impress that Levi Waldron gave one of the best talks they’d ever seen. The paper isn’t up yet (I think) but here is the R package with the data he described. His survHd package for fast coxph fits (think rowFtests but with Cox) is also worth checking out.
- John Cook kept cranking out interesting posts, as usual. One of my favorites talks about how one major component of expertise is the ability to quickly find and correct inevitable errors (for example, in code).
- Larry Wasserman’s Simpson’s Paradox post should be required reading. He is shutting down Normal Deviate, which is a huge bummer.
- Andrew Gelman and I don’t always agree on scientific issues, but there is no arguing that he and the stan team have made a pretty impressive piece of software with stan. Richard McElreath also wrote a slick interface that makes fitting a fully Bayesian model match the syntax of lmer.
- Steve Pierson and Ron Wasserstein from ASA are also doing a huge service for our community in tackling the big issues like interfacing statistics to government funding agencies. Steve’s Twitter feed has been a great resource for keeping track of deadlines for competitions, grants, and other deadlines.
- Joshua Katz built these amazing dialect maps that have been all over the news. Shiny Apps are getting to be serious business.
- Speaking of RStudio, they keep rolling out the goodies, my favorite recent addition is interactive debugging.
- I’ll close with David Duvenaud’s HarlMCMC shake: